Leukemia (1997) 11, 1124–1130 1997 Stockton Press All rights reserved 0887-6924/97 $12.00 Flow cytometric analysis of P-glycoprotein function using rhodamine 123 JPe´triz and J Garcı´a-Lo´pez Departament de Criobiologia i Tera`pia Cel·lular, Centre de Refere`ncia de Citometria Izasa-Coulter, Institut de Recerca Oncolo`gica, Hospital Duran i Reynals, Gran Via Km 2,7, L’Hospitalet de Llobregat, 08907, Barcelona, Spain The MDR1 gene product, P-glycoprotein (P-gp), works as a ing its functional activity.24–26 The detection of P-gp activity transmembrane efflux pump for several cytotoxic products, in human tumor samples is of great clinical interest, parti- representing a major cause for cancer treatment failure. Rhoda- cularly for investigating certain agents that are known to mine 123 (Rh123), a low toxic fluorescent probe commonly 27,28 used to assess mitochondrial bioenergetics in living cells, has reverse the multidrug resistant phenotype. Functional also been used to measure the efflux activity of P-gp in both analysis of the P-gp can be accomplished using diverse normal and malignant cells. Analysis of variation in cellular techniques: (1) transport experiments with radiolabeled fluorescence by measuring the rates of Rh123 influx and efflux, substrates29 (ie 3H vinblastine); (2) analysis of drug transport together with the effect of mdr reversing agents, allows the kinetics using P-gp substrates attached to fluorescent com- investigation of drug-resistant phenotypes in cancer samples. pounds30 (ie coumarin-vinblastine, photoactivatable taxol); We have studied the functional activity of P-gp in human leu- 31 kemic cell lines using flow cytometry, taking into consideration and (3) flow cytometric efflux assays with fluorescent dyes. that variables such as Rh123 cytotoxicity, culture conditions, For this last purpose, and since certain anticancer agents such cell membrane integrity, as well as the effect of specific P-gp as anthracyclines32 (doxorubicin, daunorubicin) have fluor- modulators, can impair the resolution of the Rh123-efflux escent properties, they can be used to quantitate their cellular measurements. The studies show that: (1) optimal non-cyto- influx, retention and efflux in tumor cell populations. Some toxic concentrations of Rh123 which allow appropriate color other relatively non-toxic dyes are well known substrates for compensation are in the range of 50–200 ng/ml; (2) life-gating 33 allows accurate measurement on the 50% average rate of P-gp, such as rhodamine-123 (Rh123), bisbenzimidazole 34 Rh123 efflux; (3) relative efficiency of P-gp inhibitors was PSC- dyes (Hoechst 33342), 3,3′-diethyloxacarbocyanine iodide . ′ 833 cyclosporin A verapamil; and (4) the presence or (DiOC2(3)), 3,3 -diethyloxadicarbocyanine iodide (DODC absence of fetal calf serum had no effect on the bioavailability iodide),35 thiazole and oxazole derivatives36 (SY-38 and SY- of chemosensitizer agents, with the exception of serum-free 3150 respectively), and fluorescent calcium probes37 (ie indo- experiments, which showed a significant decrease in P-gp activity under the presence of PSC-833 (P 5 0.05). Hence, we 1AM, fluo-3AM). These substrates can also be used, together recommend this experimental strategy for clinical practice bet- with P-gp inhibitors, to study MDR functionality. Rh123 is a l = ter to study the cellular drug resistance phenotype. cationic and lipophilic fluorescent dye ( ex/em 505/534 nm) Keywords: P-glycoprotein; multidrug resistance; rhodamine 123; more photostable than fluorescein, pH insensitive under flow cytometry physiological conditions, and membrane permeable, which shows a rapid uptake by mitochondria in living cells.38 It has been used in diverse studies, particularly in the investigation Introduction of drug resistance,39 assessment of mitochondrial bioenerget- ics in living cells40 (ie enzymatic activity and mitochondrial Multidrug resistance (MDR) in tumor cells is associated with transmembrane potential assessment); isolation of primitive the overexpression of mdr encoding genes, whose expression bone marrow hematopoietic progenitor cells,41 the study of confers cellular drug resistance against some specific anti- cell-cycle processes,42 and the staining of living micro- tumor agents.1–4 P-glycoprotein (P-gp), the product of the mdr organisms.43 With regard to drug resistance assessment, con- gene, is a Mr 170 000–180 000 energy-dependent transmem- ditions for the action of P-gp depend on the mitochondrial brane pump5–7 which extrudes a diverse group of unrelated membrane potential and cell membrane potential, as well as compounds, such as the Vinca alkaloids, anthracyclines, epi- on the time and temperature of exposure, and drug concen- podophyllotoxins, taxol, and certain protein synthesis inhibi- tration. Since living cells are sensitive to these changes,44 pre- tors,8–11 resulting in a decreased, and less toxic, intracellular liminary studies of the transporter activity are needed in order concentration of anticancer agents.12,13 However, increased accurately to design and interpret future experiments. This levels of P-gp are not the only cause of MDR; other mech- paper focuses on the study of the P-gp function, as determined anisms can be involved, such as the overexpression of the with Rh123 and flow cytometry, as well as on the effect of drug resistant MRP (MDR-associated protein) gene, which has diverse efflux blockers on Rh123 retention by tumor cells. been observed in non-P-gp cell lines selected for resistance to cytotoxic drugs.14,15 P-gp is also expressed in a wide variety of normal cells, Materials and methods specially in secretory tissues, adrenal cortex, liver, jejunum, colon, kidney, certain capillary endothelia, peripheral blood Cell cultures lymphocytes, and hemopoietic precursor cells.16–21 P-gp lev- els can be analyzed by identifying the protein with specific KG1, KG1a, and K562 cell lines were obtained from the antibodies22 (ie immunohistochemistry, Western blot, flow American Type Culture Collection (Rockville, MD, USA) and cytometry), by measuring the MDR1 mRNA levels23 (ie reverse grown in RPMI 1640 nutrient mixture (Imperial, Andover, transcriptase-PCR, Northern blot), and, in addition, by analyz- UK), supplemented with 2 mML-glutamine, 1 mm sodium pyruvate, 100 units/ml penicillin, 100 mg/ml streptomycin (Biological Industries, Kibbutz Beth Haemek, Israel), and 10% Correspondence: J Pe´triz fetal calf serum (FCS). Cells were cultured at 37°C in a humidi- Received 3 October 1996; accepted 20 December 1996 fied atmosphere of 5% CO2, in air. Exponentially growing Functional expression of P-glycoprotein JPe´triz and J Garcı´a-Lo´pez 1125 cells with an optimal density of 5 × 105 cells/ml were prepared to the culture medium, in the presence or absence of either for uptake and retention experiments. Cell lines were period- Vpl, CsA, or PSC-833. After 1 h of incubation, cells were ically tested for Mycoplasma infection and found to be nega- washed and fed with Rh123-free culture medium, and cul- tive. Cell viability was determined by trypan blue exclusion tured for 90 min at 37°C, again in the presence or absence of using a hemocytometer. the MDR reversing agents, to evaluate their effect on Rh123 retention. Afterwards, cells were incubated in Rh123-free medium. Life-gate was based in light scatter parameters and in Clinical samples simultaneous staining with 5 mg/ml propidium iodide. Cellular efflux of Rh123 was measured by monitoring its fluorescence Peripheral blood and bone marrow specimens from healthy decrease at 525 nm emission wavelength. Analysis of 104 cells donors were collected in heparinized tubes. Mononuclear per sample was carried out in the Rh123/count four decades cells were prepared by Ficoll–Hypaque density gradient cen- log histogram, collecting the autofluorescence signal in the trifugation (Pharmacia, Uppsala, Sweden) according to the first decade. All analyses were performed in triplicate in three manufacturer’s recommendations, washed, and resuspended separate experiments, and the results were expressed as the in phosphate-buffered saline (pH = 7.4). In all cases viability mean of fluorescence intensity. was .90% by trypan blue exclusion. Multicolor flow cytometric analysis Source of drugs and chemicals Single suspensions of Rh123 loaded cells were incubated in Rhodamine 123 (Rh123) was obtained from Lambda Fluoresz- the presence of 20 mg/ml of CD34 monoclonal antibody in enztechnologie, (Graz, Austria); propidium iodide (Pl) from PBS-BSA (HPCA-2 PE; Becton Dickinson, San Jose, CA, USA). Sigma Chemical (St Louis, MO, USA); 7-aminoactynomycin After incubation, cells were stained with either 5 mg/ml of Pl D from Molecular Probes Europe (Leiden, The Netherlands); or 2 mg/ml of 7-AAD, 10 min prior to analysis. verapamil (Vpl) from Laboratorios Knoll, Spain; Cyclosporin A (CsA) and PSC-833 from Sandoz Pharma (Basel, Switzerland). Flow cytometry Rhodamine 123 uptake/retention assays Flow cytometric data were collected on an EPICS Elite ESP cell sorter and on an EPICS XL-MCL (Coulter Electronics, In order to achieve optimal growth conditions, cells were Hialeah, FL, USA). All studies were carried out using an air- seeded 12 h before the Rh123 uptake/retention experiments cooled 15 mW argon laser (Cyonics, Hialeah, FL, USA), in 6-well tissue culture plates (Costar, Cambridge, MA, USA), operating at a 488 nm wavelength. Forward scatter and side at a final concentration of 5 × 105 cells/ml, in 3 ml of RPMI scatter were collected on linear scale and were used to 1640 media (Gibco, Life Technologies, Gaithersburg, MD, exclude dead cells and cell aggregates either with logarithmi- USA). Rh123 uptake assays were performed by adding cally amplified Pl fluorescence (collected through a 620 nm 200 ng/ml of the dye (stock solution 1 mg/ml in distilled H2O) bandpass filter) or with 7-AAD fluorescence (collected through a 675 nm bandpass filter). Analyses were carried out using the Elite Workstation Software Ver. 4.01 and the WinMDI Ver 2.1.3. Software. Results and discussion Rh123 uptake and retention depends on both the mitochon- drial and the cell membrane potential.